React, Node, and Machine Learning - Suz Hinton @ ReactNYC

Published: 06 November 2017
on channel: React NYC

708

Suz will give a basic overview of what machine learning is, what it's used for, and how you can use NodeJS to train your own models to be only an API call away.
CONTENTS

0:05 - Into
0:33- About Suz
1:17 - What is machine learning?
3:50 - What are common uses for machine learning?
3:57 - Prediction, classification, clustering, anomaly detection
4:34 - How does machine learning work?
4:59 - Let’s look at neural networks
5;14 - Perceptron
7-16 - Neural network
7:57 - Convolutional neural net?
8:04 - Let’s start with an example
11:13 - “Make your own neural network” by Tariq Rashid
11:34 - Brain.js
12:08 - Can i just have API
13:25 - Computer Vision Service, Custom Vision Service, Custom Speech Service
13:54 - Am I stuck with the default models?
14:15 - “Who is it?” demo
15:40 - React, Redux, Redux Saga, Express
16:17 - Pro-tip: choose the right model
16:30 - Custom Vision Service Model
17:12 - Train a face model for faces
17:23 - What about speech?
19:29 - Can we use machine learning?
19:45 - Maybe, but we need lots of training data!
19:59 - How can we produce it?
21:57 - Translator API + Custom Speech Service
22:49 - I wrote some NodeJS stuff for you to use
23:34 - On more serious note
24:19 - Want to learn more?
24:56 - Questions

RESOURCES

YouTube channel
   / suziam

Twitch channel
  / noopkat

YouTube

Acoustic Model Machine
https://github.com/noopkat/acoustic-m...

Microsoft Translator Speech Service
https://github.com/noopkat/ms-transla...

Rachel Thomas
   / @math-rachel

Gene Kogan
https://ml4a.github.io

Daniel Shiffman
   / shiffman

Microsoft Cognitive Services
https://azure.microsoft.com/en-us/ser...

On this page of the site you can watch the video online React, Node, and Machine Learning - Suz Hinton @ ReactNYC with a duration of hours minute second in good quality, which was uploaded by the user React NYC 06 November 2017, share the link with friends and acquaintances, this video has already been watched 708 times on youtube and it was liked by 17 viewers. Enjoy your viewing!